This AI helps detect wildlife health issues in real time
during the spring, A disturbing pattern occurs as seabirds die along the California coast from domoic acid poisoning, caused by harmful algal blooms. Early evidence points to when and where this problem begins to spread: California’s rescued brown pelicans, red-throated pelicans, and other species begin showing up at wildlife rehabilitation centers with signs of neurological disease. However, although they populate the state map, these centers are not interconnected enough to nip the problem in the bud. When staff at one center diagnose a sick bird, others 40 miles down the road may not be privy to this information.
So researchers at the University of California at Davis recently tested an early detection system that uses artificial intelligence to rank admissions to rehabilitation centers, hoping to send warnings from wildlife agencies and researchers about the growing problems among seabirds and many other species of animals. Their system scans intake reports issued at 30 centers in California, listing information such as animal type, age, reason for admission, and diagnosis. The AI then uses natural language processing to categorize the reports, looking for patterns in the number of entries related to specific diseases and injuries.
The researchers used five years of data and more than 200,000 records to create baselines for how frequently these conditions occur naturally. When the system detects an anomaly – an unusually high number of cases in a particular species – it automatically creates an alert that is delivered to wildlife experts either via the system dashboard, email or text message. Because the system processes rehab admission data in just a day or two, it can issue “pre-diagnose” alerts, which are faster than waiting for a diagnosis to be confirmed.
In July, the team published a paper describing testing their system in the journal Proceedings of the Royal Society. says Devin Dombrowski, Head of the Wild Neighbors Database Project and one of the paper’s authors.
During the one-year pilot study, the system identified several patterns that indicate larger problems. The influx of seabirds with neurological symptoms such as head trembling and whole body trembling triggered an alert. Upon postmortem examination, it was found that these birds, including white and black waterfowl species, had domoic acid poisoning. A few months ago, the high rate of admissions to clinics in the San Francisco Bay Area for rock pigeons that showed symptoms of a neurological disease led to another alert. Further investigation proved the parasite Sarcocystis calchasi as a reason.
Study co-author Tera Kelly, a veterinarian and epidemiologist at the University of California at Davis, compares the system to syndromic monitoring of people, which uses electronic health records to monitor public health concerns, such as influenza outbreaks, opioid overdoses, and the spread of Zika and COVID-19. She points out that an animal alert system could benefit human health, too. “Wild animals could be an early indicator” of diseases such as West Nile virus, she says. The disease, which has killed more than 2,000 people since 1999, according to the Centers for Disease Control and Prevention, is often first noticed in sick birds before being diagnosed in pets and humans.
Additionally, Kelly says, “We can detect the first animal of an invasive species to be introduced to a center in California.” For example, if the numbers of mourning pigeons that were accepted into wildlife centers suddenly changed, the system could create an alert that would signal to veterinarians that the Eurasian pigeon had arrived; It is an invasive species that competes for food and can spread parasites to domestic doves.